This study provides an analysis of public debt dynamics with a view to assess the optimal size and composition of public debt in Zimbabwe that is consistent with maintaining public debt at sustainable levels. The analysis was performed by applying public debt data for Zimbabwe over the period 1980 to 2012. Robustness checks were conducted, using data for selected low income countries in the sub-Saharan Africa. The study was motivated by the public debt management concerns caused by the 2008/09 global financial crisis and the European sovereign debt crisis as well as the external public debt overhang experienced by Zimbabwe since the year 2000. The findings of the study complement existing research findings and information on public debt management of the International Monetary Fund (IMF) and other researchers. The major contribution of this thesis is the determination of optimal public debt thresholds for Zimbabwe. The optimal public debt thresholds were estimated from a joint analysis of the macroeconomic variables that affect public debt and the reaction of fiscal policy to changes in debt. The classical linear regression and Bayesian Vector Auto Regression (BVAR) models were applied to examine the drivers of debt accumulation and to assess the sensitivity of debt to macroeconomic shocks. The information from the drivers of public debt accumulation, together with the fiscal response mechanism was used to calibrate the long-run stable (optimal) public debt target. The optimal public debt threshold was also determined by assessing the link between public debt and economic growth. This assessment was carried out to establish the tipping point beyond which public debt adversely affects growth. Such a tipping point provides valuable information on the optimal size of public debt. The study also applied simulation approaches to determine the optimal composition of public debt. The results show that public debt dynamics in Zimbabwe largely comprised extensive stock flow adjustments emanating from extra budgetary expenditures to meet social and political related needs. The results of the assessment on the sensitivity of public debt to macroeconomic shocks show that Zimbabwe‟s public debt has been more vulnerable to economic growth, exchange rate and interest rate shocks. The significant influence of these variables highlights the role of automatic debt dynamics in public debt management. The results from the fiscal reaction function show that government has been responding positively to increases in public debt. This analysis also shows that government‟s policies are a-cyclical; as explained by the negative and insignificant response of the primary balance to the output gap. The dynamic stochastic simulation analysis suggests that Zimbabwe‟s public debt could follow an array of potential paths depending on the policy stance implemented by government. The simulated risk to public debt dynamics is larger, with an upper bound public debt to GDP ratio of 100 per cent and a lower bound public debt ratio of 32 per cent. The simulated lower bound provides a measure of a natural debt limit, which the government could adopt without fearing the risk of default. The results suggest that the main risks to public debt sustainability lie in growth shocks, whose volatility have been high for the period under study. The results from the analysis of growth and debt confirm the existence of an optimal growth maximising public debt ratio depicted by an inverted U-shaped relationship between public debt and economic growth. The optimal size of public debt was found to be at public debt levels of between 45-50 per cent of GDP. This means that higher public debt ratios have been associated with lower economic growth rates at debt levels above 50 per cent of GDP. The results are consistent with empirical findings for low income countries which suggest the existence of a debt laffer-curve. The results from an analysis of an optimal composition of public debt show a trade-off between a debt composition with more external concessional debt and one with more domestic debt. While a composition with more concessional borrowing was found to be desirable from a cost perspective, it proved to be less desirable from a risk perspective after taking into consideration stock flow adjustments due to changes in cross exchange rates. The findings of the study point to a need for the Zimbabwean government to swiftly respond to increases in public debt to control the swings in debt dynamics caused by macroeconomic shocks. The inverted U-shaped relationship between debt and growth suggests that government borrowing must be done in a way that simultaneously entrenches debt sustainability and ensures sustained economic growth rates in the medium to long-term. The study also highlights the need for counter-cyclical macroeconomic policies to avoid explosive debt dynamics emanating from frequent changes in the business cycle, and to minimise the interest/growth rate differential to ensure sustainable public debt dynamics. There is also a need for authorities to ensure a true balance between external and domestic borrowing to minimise the volatility in debt service costs caused by macroeconomic shocks. Generally, the findings from this study can assist in informing the policy agenda to address the imperatives of debt resolution, fiscal consolidation and economic growth acceleration.

This study examines the public debt, public debt service and economic growth nexus in Zambia, Zimbabwe and South Africa using time-series data from 1970 to 2017. This research provides empirical evidence to contribute, firstly, to the ongoing public policy debate regarding the dynamic relationship between public debt, public debt service and economic growth, and their causal relationship; and secondly, to the relative impact of domestic and foreign public debt on economic growth in the selected study countries. For this purpose, four empirical models were utilised and estimated using the Autoregressive Distributed Lag (ARDL) bounds to cointegration and the error correction ARDL-based causality test. Model 1 explored the impact of aggregate public debt on economic growth, while Model 2 investigated the relative impact of domestic and foreign public debt on economic growth. Model 3 examined the impact of public debt service on economic growth, whereas the causality between aggregate public debt and economic growth, and between public debt service and economic growth is tested in Model 4a and Model 4b, respectively. Results show that in Model 1, aggregate public debt has a positive impact on economic growth in Zambia but is negative in Zimbabwe and South Africa. In Model 2, domestic public debt negatively impacts economic growth in Zambia and Zimbabwe and positive impact in South Africa. In addition, foreign public debt has a positive impact on economic growth in Zambia and negative impact in Zimbabwe and South Africa. The results from Model 3 largely support a negative relationship between public debt service and economic growth in Zambia and Zimbabwe, and an insignificant relationship in South Africa. The causality results for Model 4a indicate that it is economic growth that drives public debt in all the study countries. Finally, no causal relationship between public debt service and economic growth was confirmed in all the study countries (Model 4b). / Economics / D. Phil. (Economics)